Study Of Mine Gas Pre-warning Based On Data Integration Mode Of Cloud Computing

As one of the major hazards, mine gas seriously impedes safe production in coal mines.The study on effective analysis of mine monitoring data for the pre-warning of abnormal gassituation is important means to expand functions of the safety monitoring and control systemas well as to improve safety pre-warning. Based on the monitoring data and data integration,mine gas pre-warning under cloud computing environment is studied.Characteristics of gas monitoring data and its integration models are studied. The timesequence of gas monitoring data is re-built in a smooth process on the basis of statistics so asto treat data defects caused by sensor errors and irregularity of measurement due to variousinterferences from environment, human and management factors, thus the data is ready forcloud computing.Mathematical models of gas concentration prediction and pre-warning of abnormal gasemission is studied. By performing data pre-processing and the ARMA time sequenceanalysis, a dynamic prediction and pre-warning model for the monitoring system is putforward, and the on-line dynamic pre-warning is realized based on real-time prediction of gasconcentration and statistics of gas measurements.Hazard prediction and pre-warning of mine gas outburst is studied. By identifying thecharacteristics of monitoring data, the progressive trend and the drift rate of gas concentration,features of gas emission, and the risks of gas outburst can be judged, and gas outburst hazardprediction and pre-warning models based on v-SVM platform is constructed, and on-site riskprediction of gas outburst is realized.A model of mine gas pre-warning supported by cloud computing is studied, and thephysical structure and platform model of cloud computing suitable for gas pre-warninganalysis is put forward. Then the algorithms of data processing and gas pre-warning areencapsulated, and services of cloud computing as well as effective pre-warning analysis arerealized. The gas pre-warning under cloud computing environment is studied. Based on gaspre-warming models, the pre-warning analysis supported by cloud computing is adopted andpromoted for on-site application. Verified by systematically obtained monitoring data, itshows that the models exhibit good applicability and validity.Therefore, this study of gas pre-warning analysis theory and methods under cloudcomputing environment is applicable to mine site. It is capable to offer scientific thinking forthe digital platforms of mine gas disaster prevention and control.